Popsom7

A User-friendly Implementation of Kohonen's Self-Organizing Maps

An implementation of Kohonen's self-organizing maps with a number of distinguishing features:

1. An easy to use interface for building and evaluating self-organizing maps.

2. Automatic centroid detection and visualization using starbursts.

3. Two models of the data: (a) a self organizing map model, (b) a centroid based clustering model.

4. A number of easily accessible quality metrics for the self organizing map and the centroid based cluster model.

Other documents:

• For a worked Python example check this notebook on Kaggle.

• For an example in R check this notebook on Kaggle.

• The Python documentation can be found here.

• For the R documentation as part of CRAN check here.